aws-samples/sagemaker-cv-preprocessing-training-performance

SageMaker training implementation for computer vision to offload JPEG decoding and augmentations on GPUs using NVIDIA DALI — allowing you to compare and reduce training time by addressing CPU bottlenecks caused by increasing data pre-processing load. Performance bottlenecks identified with SageMaker Debugger.

30
/ 100
Emerging

This project helps machine learning engineers and data scientists accelerate computer vision model training on Amazon SageMaker. It takes your existing image datasets and training scripts and outputs faster training times and insights into CPU bottlenecks. This is especially useful for those working with large image datasets and complex augmentation pipelines.

No commits in the last 6 months.

Use this if you are a machine learning engineer experiencing slow training times for your computer vision models on Amazon SageMaker due to heavy image preprocessing and augmentations.

Not ideal if your training bottlenecks are not related to CPU-bound image decoding and augmentation, or if you are not using Amazon SageMaker for your model training.

computer-vision machine-learning-engineering model-training performance-optimization image-preprocessing
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 6 / 25
Maturity 16 / 25
Community 8 / 25

How are scores calculated?

Stars

21

Forks

2

Language

Python

License

Last pushed

Jul 13, 2021

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/aws-samples/sagemaker-cv-preprocessing-training-performance"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.